PGH_Chrispy
Smash Journeyman
Hello! I've been looking at the recent thread on the released Japanese tier list, as well as looking at the controversy over the use of tiers in general, and have begun to think about different methods that could be used to rank characters.
One thing I noted was that, in general, the problem with tier lists is that the highest ranked characters end up being the ones that the most people start playing, leaving lower tiers stagnate. Instead of lamenting this fact, why not make it a feature?
My essential idea is to make an "Ease of Use" tierlist, where characters are ranked by how easy they are to pick up and become good at. This would serve to give people with limited time an idea of who they could reasonably learn. The details of this are still somewhat ambiguous, but the tierlist would be formed with level 9 CPU clones as the baseline. After a short amount of preparation, enough to understand the moveset and some potential advanced techniques, testers would go into battle against the CPU of their character; a good performance would indicate a character that is easy to learn, while a poor performance would suggest that the character is complex and requires more input to be expertly manipulated.
How exactly to conduct the battles is pretty open to interpretation. It would be possible to do everything from timed matches, stocks from 1 to 99, best out of x sets, and play until you win so many times, potentially in a row, as well as items and stage choices. Scoring could be based on percentages dealt or total KO's.
To test the viability of this, I decided on a single 7 stock match, no items, Battlefield, against a level 9 CPU clone, and tested the first 6 characters - Mario, Luigi, Peach, Bowser, Yoshi, and Roselina. All of these characters are characters I've had a small amount of experience with in SSB4, with Peach and Bowser having more experience from previous games. Scoring would be based on how many stocks were left at the end - a positive number indicates victory, a negative indicates a loss.
Of course, this is not a full picture of these characters and how easy they are to learn. One match is not enough to know about one character, and one person testing, especially, is not the best way to test. Were I to do the full roster, the scores of my mains such as Zelda would be severely skewed. Thus, a large sample size would need to be procured, with the proper statistical method, to end up with a tier list truly representative of how easy it is to learn a character.
One final thought: This list would not be intended to fully replace traditional tier lists. Those lists measure competitive placement, and still serve that purpose. This list merely attempts to separate ease of use from competitive viability, a distinction that has been all too murky in past games.
Thank you for reading my ideas! I welcome suggestions on how to implement this, criticisms on where this plan falls short, and your own statistics and ideas on what they mean.
One thing I noted was that, in general, the problem with tier lists is that the highest ranked characters end up being the ones that the most people start playing, leaving lower tiers stagnate. Instead of lamenting this fact, why not make it a feature?
My essential idea is to make an "Ease of Use" tierlist, where characters are ranked by how easy they are to pick up and become good at. This would serve to give people with limited time an idea of who they could reasonably learn. The details of this are still somewhat ambiguous, but the tierlist would be formed with level 9 CPU clones as the baseline. After a short amount of preparation, enough to understand the moveset and some potential advanced techniques, testers would go into battle against the CPU of their character; a good performance would indicate a character that is easy to learn, while a poor performance would suggest that the character is complex and requires more input to be expertly manipulated.
How exactly to conduct the battles is pretty open to interpretation. It would be possible to do everything from timed matches, stocks from 1 to 99, best out of x sets, and play until you win so many times, potentially in a row, as well as items and stage choices. Scoring could be based on percentages dealt or total KO's.
To test the viability of this, I decided on a single 7 stock match, no items, Battlefield, against a level 9 CPU clone, and tested the first 6 characters - Mario, Luigi, Peach, Bowser, Yoshi, and Roselina. All of these characters are characters I've had a small amount of experience with in SSB4, with Peach and Bowser having more experience from previous games. Scoring would be based on how many stocks were left at the end - a positive number indicates victory, a negative indicates a loss.
- Mario: -2
- Luigi: -4
- Peach: -2
- Bowser: +3
- Yoshi: +2
- Roselina: +3
Of course, this is not a full picture of these characters and how easy they are to learn. One match is not enough to know about one character, and one person testing, especially, is not the best way to test. Were I to do the full roster, the scores of my mains such as Zelda would be severely skewed. Thus, a large sample size would need to be procured, with the proper statistical method, to end up with a tier list truly representative of how easy it is to learn a character.
One final thought: This list would not be intended to fully replace traditional tier lists. Those lists measure competitive placement, and still serve that purpose. This list merely attempts to separate ease of use from competitive viability, a distinction that has been all too murky in past games.
Thank you for reading my ideas! I welcome suggestions on how to implement this, criticisms on where this plan falls short, and your own statistics and ideas on what they mean.